© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 22227
    Prev
    Next
    INDEX
    Explanations

    verbs related to physical actions or commands that involve pressing a button or key

    oai_token-act-pair · gpt-3.5-turbo

    instructions or commands related to pressing buttons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Trayvon
    -0.69
    BuyableInstoreAndOnline
    -0.67
    abama
    -0.64
     Sheldon
    -0.64
     Cloak
    -0.64
    beard
    -0.61
     Daredevil
    -0.61
    cknowled
    -0.59
     Equality
    -0.59
    iland
    -0.58
    POSITIVE LOGITS
    urized
    1.18
    press
    0.81
     presses
    0.80
    ured
    0.78
    ures
    0.77
    ature
    0.77
    roth
    0.76
     KP
    0.75
    ure
    0.73
    ername
    0.72
    Activations Density 0.016%

    No Known Activations